Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

CI: Update cudf python to 0.16 nightly #790

Merged
merged 1 commit into from
Sep 18, 2020
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
25 changes: 11 additions & 14 deletions jenkins/Dockerfile.integration.centos7
Original file line number Diff line number Diff line change
Expand Up @@ -23,6 +23,10 @@ ARG CUDA_VER=10.1

FROM nvidia/cuda:${CUDA_VER}-runtime-centos7

ARG CUDA_VER=10.1
ARG CUDF_VER
ARG URM_URL

#Install java-8, maven, docker image
RUN yum update -y && \
yum install -y centos-release-scl && \
Expand All @@ -31,27 +35,20 @@ RUN yum update -y && \
# The default mvn verision is 3.05 on centos7 docker container.
# The plugin: net.alchim31.maven requires a higher mvn version.
ENV MAVEN_HOME "/usr/local/apache-maven-3.6.3"
ARG URM_URL
RUN wget ${URM_URL}/org/apache/maven/apache-maven/3.6.3/apache-maven-3.6.3-bin.tar.gz -P /usr/local && \
tar xzvf $MAVEN_HOME-bin.tar.gz -C /usr/local && \
rm -f $MAVEN_HOME-bin.tar.gz

ENV PATH "$MAVEN_HOME/bin:$PATH"

RUN wget --quiet https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O ~/miniconda.sh && \
/bin/bash ~/miniconda.sh -b -p /opt/conda

ENV PATH="/opt/conda/bin:${PATH}"
/bin/bash ~/miniconda.sh -b -p /opt/conda && \
rm -f ~/miniconda.sh

RUN . /opt/conda/bin/activate && \
conda init && \
conda --version
ENV PATH="/opt/conda/bin:$MAVEN_HOME/bin:${PATH}"

ARG CUDA_TOOLKIT_VER=10.1
RUN conda install -y -c rapidsai -c nvidia -c conda-forge -c defaults cudf=0.15 python=3.7 cudatoolkit=${CUDA_TOOLKIT_VER} && \
conda install -y spacy && \
python -m spacy download en_core_web_sm && \
conda install -y -c anaconda pytest requests pandas pyarrow && \
# 'pyarrow' and 'pandas' will be installed as the dependencies of cudf below
RUN conda install -y -c rapidsai -c rapidsai-nightly -c nvidia -c conda-forge -c defaults cudf=${CUDF_VER} python=3.7 cudatoolkit=${CUDA_VER} && \
conda install -y spacy && python -m spacy download en_core_web_sm && \
conda install -y -c anaconda pytest requests && \
conda install -y -c conda-forge sre_yield && \
conda clean -ay

Expand Down
4 changes: 3 additions & 1 deletion jenkins/Jenkinsfile.integration
Original file line number Diff line number Diff line change
Expand Up @@ -58,6 +58,8 @@ pipeline {
agent { label 'docker-gpu' }
steps {
script {
def CUDF_VER=sh(returnStdout: true,
script: '. jenkins/version-def.sh>&2 && echo -n $CUDF_VER') - "-SNAPSHOT"
def CUDA_NAME=sh(returnStdout: true,
script: '. jenkins/version-def.sh>&2 && echo -n $CUDA_CLASSIFIER | sed "s/-/./g"')
def IMAGE_NAME="$ARTIFACTORY_NAME/sw-spark-docker/plugin:it-centos7-$CUDA_NAME"
Expand All @@ -67,7 +69,7 @@ pipeline {
// Speed up Docker building via '--cache-from $IMAGE_NAME'
def buildImage=docker.build(IMAGE_NAME,
"-f jenkins/Dockerfile.integration.centos7 --build-arg CUDA_VER=$CUDA_VER \
--build-arg URM_URL=$URM_URL --cache-from $IMAGE_NAME -t $IMAGE_NAME .")
--build-arg URM_URL=$URM_URL --build-arg CUDF_VER=$CUDF_VER --cache-from $IMAGE_NAME -t $IMAGE_NAME .")
def buildImageID=sh(returnStdout: true, script: "docker inspect -f {{'.Id'}} $IMAGE_NAME")
if (! buildImageID.equals(urmImageID)) {
echo "Dockerfile updated, upload docker image to URM"
Expand Down
2 changes: 2 additions & 0 deletions python/rapids/worker.py
Original file line number Diff line number Diff line change
Expand Up @@ -52,6 +52,8 @@ def initialize_gpu_mem():
"`RAPIDS_POOLED_MEM_SIZE`.")
if pool_max_size == 0:
pool_max_size = max_size
pool_max_size = pool_max_size >> 8 << 8
pool_size = pool_size >> 8 << 8
jlowe marked this conversation as resolved.
Show resolved Hide resolved
print("DEBUG: Pooled memory, pool size: {} MiB, max size: {} MiB".format(
pool_size / 1024.0 / 1024,
('unlimited' if pool_max_size == max_size else pool_max_size / 1024.0 / 1024)))
Expand Down